STELAR THEORIES 337 



a xylem sheath." This is, of course, possible, but it does not 

 appear altogether probable. 



In a great many roots, at any rate, the endodermis does form 

 a continuous sheath around the vascular cylinder, and even in 

 those cases where marked passage-cells are present (distinguish- 

 able by the fact that their basal walls are not thickened like the 

 other cells of the endodermis), the radial walls are often 

 cuticularised. 1 Then again, although the xylem of aquatic plants 

 is much reduced it may not be entirely absent, and it performs 

 its normal function of water-carriage. This fact was shown by 

 Sauvageau 2 and, more recently, by Pond, 3 who demonstrated 

 that the plants he experimented with, viz. Vallisneria spiralis, 

 Ranunculus aquatilis trichophyllus, Elodca canadensis, Myriophyl- 

 luni spicalum, and species of Potamogcton are dependent on the 

 substratum for a considerable portion of their food-material, and 

 that there is a current of water upwards from the roots. 



Finally, the reason why the xylem requires protection against 

 the access of air is clear. There frequently exists in the tracheae 

 a marked negative pressure ; should air gain an entry into these 

 elements, while in this condition, the passage of water through 

 them would be enormously impeded, if not entirely stopped, 

 on account of the much greater resistance to the flow. 4 



On the other hand, the reason why the phloem elements 

 should require the protection of an endodermis is not obvious. 

 On the whole the anatomical evidence appears to support the 

 first possibility, considered above, which is the view held by 

 Strasburger, Boodle, and others. Jeffrey controverts this inas- 

 much as no experimental evidence has been brought forward. 



Another use assigned to this tissue is connected with the 

 reaction of certain parts of plants to the stimulus produced 

 by weight, the consideration of which is outside the scope of 

 the present paper. 5 



1 The probability is not excluded that in those cases where the endodermis is 

 greatly thickened both on its radial and basal walls, it may also perform a 

 mechanical function. 



- Sauvageau, C, Sur les fcuillcs de quelques Monocotyledones aquatiques. 

 Paris 1891. 



3 Pond, Raymond H., "The Biological Relation of Aquatic Plants to the 

 Substratum." U.S. Fish Com. Rep., 1903. 



4 See Ewart, " The Ascent of Water in Trees," Phil. Trans. Roy. Sec. Lond. 

 B. 198, 1905. 



5 See Darwin, Francis, Presidential address, British Association, Section K, 

 Cambridge, 1904. 



22 



